x-43d conceptual design and feasibility study - nasa technical

1

American Institute of Aeronautics and Astronautics

An Overview of NASA's Integrated Design and

Engineering Analysis (IDEA) Environment

Jeffrey S. Robinson*

NASA Langley Research Center, Hampton, Virginia 23681

Historically, the design of subsonic and supersonic aircraft has been divided into

separate technical disciplines (such as propulsion, aerodynamics and structures), each of

which performs design and analysis in relative isolation from others. This is possible, in most

cases, either because the amount of interdisciplinary coupling is minimal, or because the

interactions can be treated as linear. The design of hypersonic airbreathing vehicles, like

NASA’s X-43, is quite the opposite. Such systems are dominated by strong non-linear

interactions between disciplines. The design of these systems demands that a multi-

disciplinary approach be taken. Furthermore, increased analytical fidelity at the conceptual

design phase is highly desirable, as many of the non-linearities are not captured by lower

fidelity tools. Only when these systems are designed from a true multi-disciplinary

perspective, can the real performance benefits be achieved and complete vehicle systems be

fielded.

Toward this end, the Vehicle Analysis Branch at NASA Langley Research Center has

been developing the Integrated Design & Engineering Analysis (IDEA) Environment. IDEA

is a collaborative environment for parametrically modeling conceptual and preliminary

designs for launch vehicle and high speed atmospheric flight configurations using the

Adaptive Modeling Language (AML) as the underlying framework. The environment

integrates geometry, packaging, propulsion, trajectory, aerodynamics, aerothermodynamics,

engine and airframe subsystem design, thermal and structural analysis, and vehicle closure

into a generative, parametric, unified computational model where data is shared seamlessly

between the different disciplines. Plans are also in place to incorporate life cycle analysis

tools into the environment which will estimate vehicle operability, reliability and cost.

IDEA is currently being funded by NASA’s Hypersonics Project, a part of the

Fundamental Aeronautics Program within the Aeronautics Research Mission Directorate.

The environment is currently focused around a two-stage-to-orbit configuration with a

turbine-based combined cycle (TBCC) first stage and a reusable rocket second stage. IDEA

will be rolled out in generations, with each successive generation providing a significant

increase in capability, either through increased analytic fidelity, expansion of vehicle classes

considered, or by the inclusion of advanced modeling techniques. This paper provides the

motivation behind the current effort, an overview of the development of the IDEA

environment (including the contents and capabilities to be included in Generation 1 and

Generation 2), and a description of the current status and detail of future plans.

I. Introduction

N the world of conventional aircraft design, technical disciplines can operate in relative isolation from each

other because cross-discipline interactions are often either minimal or at least can be treated as linear. On the

contrary, the design of hypersonic airbreathing vehicles, like NASA’s X-43 vehicle shown in Figure 1, is

dominated by strong non-linear interactions. For instance, the forebody and aftbody surfaces on the underside of

vehicle provide the majority of the vehicle’s total aerodynamic lift, but also act as the inlet and nozzle for the

scramjet engine. As such, both the aerodynamic and propulsion disciplines are greatly affected by their design,

which is often determined through a multi-disciplinary optimization performed at the vehicle level. Such trade-offs

and multi-disciplinary analyses are common for this class of vehicle and, in fact, are required for the design to

achieve its full performance potential1.

* Aerospace Engineer, Vehicle Analysis Branch, MS 451 NASA Langley Research Center, Senior Member

I

https://ntrs.nasa.gov/search.jsp?R=20110010978 2019-04-16T10:53:23+00:00Z

2


In the United States, the hypersonics community

(government, industry and academia) strongly agrees

that the key to unlocking the potential in hypersonic

aircraft lies in multi-disciplinary analysis at the vehicle

level and that improvements in this capability are

critical to future success. In 2005, at the request of the

United States Congress, the National Institute of

Aerospace (NIA) developed and released “Responding

to the Call: Aviation Plan for American Leadership”2, a

1000+ page document which detailed the deterioration

of America’s dominance in aviation and aeronautics

research. It provided, as a start towards recovery, a detailed plan in each of seven aeronautics sectors, among which

was hypersonics. In the hypersonics plan, the first critical area identified was Multidisciplinary Design, Analysis and

Optimization (MDAO), stating, “The highly integrated nature of hypersonic vehicles, combined with their high

levels of technological and economic uncertainty, render conventional design practices inadequate for synthesizing

systems to meet all performance, effectiveness, and economic requirements. Improved methods of system design that

account for and even take advantage of the highly integrated nature of hypersonic vehicles are therefore crucial to

their successful development.” The plan went on to describe the components and attributes of an integrated design

and optimization environment, saying that “Successful hypersonic vehicle design is not possible without such

improved, integrated and automated methods.” The need identified here by the NIA has also been detailed by the

U.S. Air Force3, Boeing

4 and NASA

5.

II. Background

Figure 2 shows the combination of analytical

disciplines typically involved in the design, analysis

and optimization of hypersonic airbreathing vehicles.

Among these ten, “Life Cycle Analysis” encompasses

an additional set of disciplines that help to provide

estimates of system cost, reliability and operability.

Classic MDAO methods (response surface fitting

techniques, multi-objective / multi-attribute optimiza-

tion, numerical smoothing, uncertainty quantification /

uncertainty propagation, etc.) are captured under

“Optimization & Advanced MDAO Techniques”. The

remaining eight discipline areas are those that are

traditionally included in determining the overall

performance of the system.

Numerous attempts have been made in the

past by NASA and others to integrate these disciplines

into a unified environment. Different frameworks with varying levels of integration have been fielded, yielding

mixed results. One of the more notable efforts in recent years was the Advanced Engineering Environment (AEE)6,

funded by NASA’s Space Launch Initiative. AEE was built utilizing Phoenix Integration’s ModelCenter©

framework. While AEE worked well for expendable and reusable rocket-based launch vehicles, it lacked the

detailed geometry capability that is crucial to accurately model and analyze hypersonic airbreathing vehicles. The

hypersonics group within Boeing recognized this need and endeavored to develop their own internal parametric

geometry modeling capability that ultimately would become the heart of their environment, BIVIDS4. Researchers

at the Air Force Research Laboratory (AFRL) also saw this need and found their answer3 with the Adaptive

Modeling Language (AML)7, a product of Technosoft, Inc. While it can communicate natively with other

commercial computer aided design (CAD) packages (Pro-E, Catia, etc.), AML and its environment, like the Boeing

system, have at its core a parametric geometry modeling capability. This feature is critical, as it allows each

discipline to natively share, understand and interpret the knowledge of the same geometry. Often, in the more typical

design cycle where the disciplines are not well-integrated, it is common for each discipline to generate its own

representation of the actual geometry, leading to potential inconsistencies and complicating configuration control.

With AML controlling and distributing information about the geometry in the form required by each discipline, this

issue is avoided. In addition to parametric geometry generation, other requirements for the environment include

Figure 2. Graphic showing analytical disciplines

involved in hypersonic systems analysis and design.

Figure 1. Artist’s concept of X-43 showing airflow

along vehicle forebody and aftbody.

3


streamlined data transfer between analysis tools; automated coupling and execution of computational analyses;

multi-disciplinary design optimization methods; and probabilistic methods and processes that enable system level

risk assessment/mitigation and robust vehicle configuration optimization. The environment must also support and

integrate multiple levels of analytical fidelity.

AFRL researchers introduced AML to the hypersonics group in the Vehicle Analysis Branch (VAB) at NASA

Langley in 1998. Since then, VAB has been partnering with Technosoft through a multi-phase Small Business

Innovative Research (SBIR) award to develop interfaces in the AML environment for some of VAB’s legacy codes8.

Initially focused on providing engineers with enhancements to their individual discipline tools, the focus has shifted

over the last several years towards integrating these tools into a unified, multi-disciplinary analysis and design

capability. Known formerly as CoHAVE and AdVISE9, the system is now referred to as the Integrated Design and

Engineering Analysis (IDEA) Environment. The current effort is being supported by the MDAO Discipline within

the Fundamental Aeronautics Program’s Hypersonics Project.

III. Discipline Fidelity Levels

During NASA’s Next Generation Launch Technology (NGLT) program, the Systems Analysis Project (SAP)

conducted and coordinated multiple sets of system analyses across various missions and with varying levels of

technology assumptions10

. In order to get a better understanding of the differences between analyses and the level of

uncertainty (generally) contained in each, the SAP endeavored to standardize definitions for the various levels of

fidelity within each of the disciplines. The MDAO Discipline within the Hypersonics Project has updated and

adopted this matrix to help guide it with tool development and as a basis for comparing analytical results on system

studies. The matrix includes five distinct levels of fidelity for the eight performance-related disciplines mentioned

previously, as well as five levels of fidelity for the disciplines that make up life cycle analysis. This matrix has also

been adopted and employed in the NASA-USAF Joint System Study, an effort established by the Air Force Chief

Scientist and NASA Associate Administrator for Aeronautics Research that endeavors to studied the application of

hypersonic airbreathing propulsion for access-to-space missions. Use of the fidelity matrix in the Joint System Study

greatly enhanced the ability of the two agencies to communicate and compare analyses of hypersonic vehicle

designs. In order to provide additional clarity, fidelity definitions shown in this paper include several updates to

those used in the Joint Sysetm Study.

The updated matrix for the performance-related disciplines is included in tables in each respective section

below. As seen in the tables, at the lowest level of fidelity (level 0), the disciplines typically employ historical or

scaled empirical data in order to estimate vehicle performance. In general, uncertainty is expected to be the highest

at this level, although computational speed and flexibility in the design space are the greatest. One can also relate the

programmatic development cycle11

and the typical system breakdown structure (SBS) or system hierarchy

(architecture > major system > element > subsystem > component > subassembly > part)12

to the various levels of

fidelity. At the beginning of any program (pre-Phase A), trade studies and systems analyses are conducted at the

highest SBS level, the architecture level. Here, the entire mission and its global requirements need to be considered

in order to determine the performance required out of each of the major systems. This program phase and SBS level

generally will incorporate analyses conducted at fidelity levels 0-1. As a program progresses into Phase A, the level

of detail in the design increases from the architecture and major system level, down to the element level. This

progression would correspond roughly with discipline analyses at fidelity level 2 and bring a design close to the

System Requirement Review (SRR) level of maturity. As the level of fidelity and the amount of detail increase, the

level of uncertainty in the design should correspondingly decrease, although computational speed continues to slow,

and design flexibility continues to become more limited. As detail increases to the subsystem and component levels,

discipline fidelity increases to levels 3 and 4, and the program pushes towards Preliminary Design Review (PDR).

At this point in the design, the majority of the design choices will have been made and standard engineering

development takes over to complete detailed subassembly and part specifications.

Within the MDAO Discipline of the Hypersonics Project, architecture level trade studies and systems analyses

are performed using the EXAMINE tool13

. EXAMINE, developed over the last five years at NASA Langley, is a

collection of Microsoft Excel©

workbooks that contain empirical data and mass estimating relationships (MERs), i.e.

data at fidelity level 0, for numerous vehicle classes and related subsystems. EXAMINE offers the ability to rapidly

perform trade studies at the architecture level to help guide major system and element requirements. The long-term

plan for IDEA is to be centered about the Level 2 fidelity capability, with the ability to run at Level 1 (and mixed

Level 1 to 2) as well as directly support analysis at Level 3 and higher. Ultimately, analyses performed at higher

levels of fidelity will be used to update or even create new models at lower levels of fidelity that are computationally

more efficient and support the cycle times needed to perform optimization at the vehicle level.

4


The current fidelity level definitions for the life cycle analysis-related disciplines (i.e. cost, reliability, and

operations) is included in that discipline’s section in Table 9. Currently, few tools exist in these areas and most are

Level 1 at best. As such, the Hypersonics Project is endeavoring to fill some of these gaps, largely through the use of

NASA Research Announcements (NRAs). One such NRA was just completed with Spaceworks, Inc., who

developed a discrete event simulation of operations for hypersonic vehicles14

. Built using Arena©, the “Descartes”

tool provides estimates of the operational characteristics of the vehicle such as turn around time and operations cost.

Ideally, the MDAO Discipline would like to issue a similar award through the NRA process for development of an

improved safety and reliability tool for hypersonic systems.

Care is being taken to ensure that IDEA can readily support analyses at higher fidelity levels. One such effort

currently underway is aimed at automated generation of structured CFD grids, guided by geometry and grid

topology, to be used with the Vulcan CFD code15

. Vulcan is a structured code that solves the full Navier-Stokes

equations for turbulent, non-equilibrium, chemically reacting flows16

. During the X-43 program, Vulcan was used to

compute full vehicle, powered solutions that were found to compare extremely well with flight data17

. Vulcan

analyses have also shown excellent agreement with powered and unpowered tests in Langley’s 8-ft. High

Temperature Tunnel, as well as other scramjet and high-speed test facilities. In 2003, Vulcan was used to compare to

a simulated powered test of a rocket-based combined cycle (RBCC) vehicle in Langley’s 16-ft transonic facility,

again with excellent agreement18

. As such, Vulcan has become the benchmark CFD tool at Langley for hypersonic

vehicles, and being able to support it directly with automated grid generation from IDEA is essential.

IV. IDEA Contents and Capabilities

The MDAO Discipline plans to develop and roll out the

IDEA environment over several generations, ultimately being

centered around the discipline tools that meet fidelity Level 2

requirements. While multiple vehicle classes will ultimately be

defined within IDEA (e.g. waveriders, “beta” boosters,

vehicles with 3-D inlets, etc.), the current environment is built

around a fully-reusable, two-stage-to-orbit (TSTO) system that

employs a turbine-based combined cycle (TBCC) lifting-body

first stage and a rocket-based winged-body second stage, as

shown in Figure 3. This vehicle served as the NASA reference

concept developed under the Joint System Study, mentioned

previously. Results from that study will be used for verification as IDEA is developed and fielded.

A. Configuration, Packaging and Geometry

The fidelity matrix for the Configuration, Packaging and Geometry discipline is shown in Table 1. For creation

of the vehicle outer mold line (OML), the geometry engine in AML will provide a fully parametric modeling

capability. Figure 4 shows the range of vehicle shapes that can be generated for the second stage with the current

class definition. The body shape is specified through an overall length, width and height at the fuselage base, and by

placing some control points along the length of the body. In general, the body has an elliptic cross-section; however,

Figure 3. Two-stage-to-orbit launch vehicle

concept with a turbine-based combined cycle

first stage and a reusable, rocket-powered

second stage.

Table 1. Fidelity level definitions for the Configuration, Packaging, and Geometry discipline.

Level 0 Level 1 Level 2 Level 3 Level 4

Parametric,

empirical or

analytical geometry

model

External & major

internal components

such as propellant

tanks payload bay,

propulsion, etc.

modeled for

volume, area, and

key linear

dimensions

Majority of

components

modeled, packaged,

and analyzed for

geometric properties

including center of

gravity. OML

includes bluntness,

surface deflection

details, etc.

All components

modeled, packaged,

and analyzed for

geometric properties

including center of

gravity and inertia

characteristics.

OML detail includes

steps and gaps, etc.

Internal components

modeled after actual

hardware elements

and real geometry.

OML detail includes

all external

protuberances.

5


Figure 6. Sample internal packaging and

structural arrangement for a second stage

concept.

options are available to flatten the sides, as seen in the X-34

design, or the vehicle bottom (making the traditional “D”

shape). Control is provided over nose bluntness and droop

angle. Wings are specified by standard parameters such as

span, leading edge sweep, chord length, etc., as well as the

overall vertical and axial wing location. The user can specify

the airfoil type, either NACA series or diamond, and can add

multiple sections over the span of the wing, controlling the

features of each section independently. The wing leading and

trailing edges can have distinct breaks at the sections, or a

smoothing feature is available that results in a wing like that

in the lower left corner of Figure 4. Tails can be added and

controlled in a similar manner.

For the booster, vehicle lofting begins with a lateral

extrusion of the high-speed keel line (development of the

keel line is discussed below in the Propulsion section). User

inputs control the maximum width of the vehicle at various

axial locations as

well as the desired

“2-dimensional” flowpath width. The effects of these controls are shown on

the three vehicles, as viewed from the bottom, in Figure 5. Wings and tails

can be added to the booster in a similar manner as the orbiter, and the user

has control over the height and axial location of maximum height of the

booster upper surface. In addition, the user has full control over the

placement of the orbiter relative to the booster. This allows control over

the mated center of gravity location and the degree to which the upper stage

is embedded into the first stage. Bluntness effects are also included on all

leading edges and can be controlled by the user. In general, such geometry

modeling capability can support fidelity levels 1-4. At lower levels of

fidelity, where more detailed geometry information is not needed, “feature

suppression” is used to mask unwanted detail.

A packaging system has also been created in IDEA that allows the user

to select from a wide range of predefined packaging items, each of which

can have an associated geometry. This geometry can be imported from

another CAD system, or can be generated from scratch using an internal

library of basic shapes. The packaging system has knowledge of the vehicle

OML geometry and thus can automatically shape packaging elements to be

conformal with the vehicle OML. This feature is quite useful when

modeling conformal fuel tanks, payload bays with doors that conform to

the OML, or when laying out structural elements (bulkheads, longitudinal beams, etc.) that conform to the vehicle’s

OML. The packaging system is also generic; the OML being

packaged can either be generated internally by IDEA or

imported from another CAD system, such as through IGES

or STEP translation. This still allows for vehicle designs

created outside of the IDEA environment to be analyzed

with IDEA. There is one drawback, however, in that an

imported geometry would not be parametric, making

modification of the design very difficult. A sample

packaging of a second stage concept is shown in Figure 6.

Here, the orbiter geometry (OML) was imported from CAD,

and the packaging system within IDEA was used to place all

of the elements shown.

For each packaging item, mass properties can be

assigned (either through a lumped amount or through an

alternate estimating method, i.e. an MER) or calculated

based on the packaging element geometry and proper

Figure 5. Various booster shapes

generated by IDEA showing effects

of vehicle and flowpath width

controls.

Figure 4. Second stage design interface showing

the variety of vehicle shapes that can currently

be modeled.

6


assignment of materials. The mass properties management system within IDEA can then easily generate integrated

mass properties for the entire vehicle. The packaging system fully supports fidelity Levels 1-4, according to Table 1,

as the primary difference between the levels is the number of packaging items that are contained within the model

(from only tanks and payload at Level 1 to the inclusion of all subsystems at Level 3 and higher). Several enhanced

features are also under development, including time dependent and trajectory dependent mass properties. Here, the

goal is to feed back time histories from the trajectory simulation of vehicle attitude, acceleration and propellant

usage to the mass properties module, in order to generate trajectory specific and time dependent propellant loading

states and corresponding mass properties.

A packaging strategy has been implemented for Generation 1. This strategy provides guidance to IDEA on the

preferred user arrangement of the main packaging elements, namely the two propellant tanks, the payload bay, and

the cockpit (if one exists). The strategy would identify the preferred order of the primary elements. The payload and

cockpit are typically defined with fixed dimensional values and are held constant with closure. Tanks are defined

such that their heights and widths are specified as percentages of the vehicle OML, allowing their lengths to vary as

dictated by propellant choices and the vehicle closure process. A typical strategy, as shown in Figure 6, is to locate

the LOX tank aft, the fuel tank forward, and place the payload bay between them. Such a strategy and parametric

definition of propellant tanks allows the total amount of propellant, as well as the ratio of propellants, to vary as

engine operational parameters are altered, vehicle performance changed, and the vehicle repackaged as closure

progresses and the vehicle is scaled.

B. Structures and Materials

Table 2 shows the analytical fidelity definitions for the Structures and Materials discipline. Separate efforts are

being undertaken to support Level 1 and Level 2 modeling. To support both efforts, a load case generation module

has been developed. This module allows the user to parametrically identify critical load cases experienced by the

vehicle for a given trajectory. Typical load cases include maximum and minimum (or maximum negative) normal

acceleration and maximum axial acceleration. The user can set the module up to automatically identify these cases

and to extract necessary information required for structural analysis such as: vehicle and propellant mass,

accelerations, and applied forces. Flight condition information will also be extracted and supplied to an aerodynamic

analysis code so that distributed aerodynamic forces or pressures can be obtained. All of this information will then

be compiled and used to develop load cases for structural analysis. Additionally, the user can specify non-trajectory-

based load cases, often used for modeling a runway bump or landing load. Here, the user is allowed to supply a

flight condition for aerodynamic analysis, if desired, as well as accelerations to be applied.

For Level 1 analysis, 1-D beam and shell theory will be used to estimate structural component masses. Here, 3-

D aerodynamic forces will be mapped to a 1-D line model, thrust and point loads applied, and a 1-D mass

distribution developed. Once section loads have been generated, a structural concept and a material system need to

be defined in order to estimate structure weights. For example, a stiffened-skin wing structure consists of skins, ribs,

and spars. A stiffened-skin fuselage would have skin, stringer, frame, and bulkhead structural elements. Section

properties are used to distribute section loads to cross-section structural elements. For instance, the longitudinal

stringers at a fuselage cross-section would be sized to resist axial force and bending moment. The skin would carry

shear and would transmit pressure loads to the stringers and frames. Many of the methods for distributing section

Table 2. Fidelity level definitions for the Structures and Materials discipline.


Parametric or

historical equations

adjusted to level 1

or higher for similar

technology and

vehicle

configuration

1D bending loads

analysis based on

structural theory of

beams, shell, etc…

with non-optimums

based on level 2 or

higher results

Limited 3D FEA

(<20,000 nodes) for

all major load cases,

structure sized to

allowables, non-

optimums

determined

empirically or

analytically

3D FEA (>20,000

nodes) for all major

load cases, structure

sized to allowables,

non-optimums

determined

empirically or

analytically.

Thermal effects

included. Dynamic

frequencies

estimated.

3D FEA (>100,000

nodes) for all load

cases, structure

sized to allowables,

non-optimums

determined

empirically or

analytically.

Thermal effects

included. Dynamic

frequencies

estimated.

7


loads to cross-section structural elements are documented in

Reference 19. From this point, shear and moment diagrams

can be created, and mass estimates for each section generated.

For Level 2 structural analysis, structural elements such as

ribs, spars, bulkheads, floors, and stringers can be created as

part of the packaging system. This allows these elements to be

conformal with the vehicle IML. Knowledge of the other

packaging elements also allows automated cutouts in the shape

of each element to be made in the structure to accommodate

them. Once the structure has been laid out, the individual

elements are sewn together and passed to Patran© or a similar

code to be meshed. A sample mesh of a second stage is shown

in Figure 7. Once the mesh is ready, it is combined with load

case information generated from the trajectory and passed to

Nastran©

to generate structural deflections. Nastran output will

then be passed to Hypersizer©

, a commercial structural sizing program from Collier Research Corporation, in order

to generate masses for each of the structural components. Several iterations of this loop will be required to generate

a final set of structural element masses, which guarantees that all bending and deformation constraints have been

satisfied. A more detailed description of the structures module has been documented separately

20. Once this sizing

system is in place, it can easily be extended to allow structural dynamics analyses, as well as analyses of hot

structures. These capabilities will likely be incorporated in Generation 3.

C. Trajectory, GNC and Simulation

For the Trajectory, Guidance, Navigation & Control (GNC) and Simulation discipline, Table 3 shows the

current fidelity definitions. The IDEA environment will employ the POST2 code21

for trajectory analysis and

vehicle simulation. POST2 is an industry standard point mass trajectory tool for simulating the motion of powered

or unpowered vehicles near an arbitrary, rotating, oblate, attracting body. POST2 can be run in various modes

encompassing levels of fidelity one through four, depending on options selected and data inputs. For IDEA, a

generalized user interface has been developed that offers full access to all inputs available in POST2. At many

points in the input setup, depending on the selection of various methods and operational flags, many of the input

variables available in POST2 become invalid. Intelligence has been added to the interface to only display those

variables and options that are valid, making the interface more user-friendly. In addition, cryptic POST2 variable

names are hidden from the user (there is a flag for experienced users that will display the variable names), and the

interface uses descriptive phrases to explain available options. All POST2 event types (primary, secondary, roving,

repeating) are accessible through the interface, providing a completely generic capability. Options to perform

automated trade studies and Monte Carlo analysis have also been incorporated. Within IDEA, the POST2 interface

has been integrated with other discipline tools to allow automated population of vehicle data into the input deck.

Output from POST2 is also used by several disciplines. Trajectory information is used to generate loads analysis

cases for structural sizing and will be used for sizing the thermal protection system (TPS) and various airframe and

engine subsystems. Propellant usage is used by the Sizing and Closure discipline to size the vehicle to a given

mission, as described below, as well as by the packaging system to generate time dependent mass properties

information.

Figure 7. Snapshot of a sample structural mesh

generated for a second stage.

Table 3. Fidelity level definitions for the Trajectory, GNC, and Simulation discipline.


Rocket equation

or energy

methods (path

following

simulation)

Optimized

ascent, flyback

& re-entry 3-

DOF point mass

simulation (un-

trimmed)

Optimized ascent,

flyback & re-entry 3-

DOF (pitch trim) point

mass simulation;

longitudinal stability

& control evaluation

Optimized ascent,

flyback & re-entry

6-DOF simulation;

longitudinal, lateral

& yaw stability &

control evaluation;

perfect GN&C

Optimized ascent, flyback

& re-entry 6-DOF

simulation; longitudinal,

lateral & yaw stability &

control evaluation; real

GN&C with gain

scheduling (or similar)

lags, noise, etc.

8


Some advanced features have been recently added to the trajectory interface, including the ability to

dynamically link one trajectory module to another, allowing trajectory branching. This feature is extremely useful in

modeling the TSTO problem. A deck modeling the orbiter ascent from stage separation to orbit can be dynamically

connected to the mated ascent simulation. Here, the orbiter simulation will take as its initial state the exact

separation conditions achieved by the mated system. With this dynamic link, changes to the mated trajectory phase

will automatically be propogated to other flight phases, and vice versa. This capability will ultimately allow trade

studies and optimization to be performed on the entire TSTO concept, with impacts from all disciplines and flight

phases included, in a more automated fashion. Trade study and optimization results should also be more accurate

when performed with this capability as a complete and exact set of state information will be transferred from flight

phase to flight phase. With so much information being passed and multiple models to update, it’s quite easy for

errors to creep into the process when these types of analyses are performed manually.

An automated “crash” recovery feature has been added that allows the user to guide POST2, during automated

execution, regarding what changes to make to the model to recover from an initial nonfeasible starting solution. This

type of situation occurs often when a large perturbation is applied to the vehicle (e.g. significant increase in mass),

and trajectory optimization is attempted with a starting solution based on the previous, non-perturbed vehicle. For

example, the orbiter ascent trajectory is typically guided by a table of pitch angle versus velocity. If vehicle mass is

increased substantially, this profile will not provide sufficient lift and upward thrust vector to allow the vehicle to

achieve orbit. It will typically crash back to Earth, and the run will terminate. The recovery feature allows the user to

link a trajectory constraint to a trajectory input parameter. For the orbiter example, the user could create a constraint

that the flight path angle at engine cutoff has to be positive. When the vehicle crashes, that constraint will not be

met. The recovery feature would allow the user to connect that constraint with the pitch angle profile. When the case

crashes, IDEA would identify that the linked constraint was not satisfied and increment the pitch profile by a user-

specified amount. This adjustment will eventually raise the flight profile enough that orbit can be attained, yielding

an initial feasible solution for optimization to begin. Additionally, an advanced run feature has been added that will

execute POST2 in targeting mode. This mode will allow the user to find a trajectory solution that satisfies all of the

constraints prior to turning on optimization. This method has been found to aid optimization in achieving a solution

more quickly.

Methods will also be incorporated at appropriate levels of fidelity that evaluate vehicle stability and control at

various points along the flight profile. Using a similar method to the structural load case generator, points along the

trajectory will be examined for longitudinal and lateral-directional static margin, control effectiveness and dynamic

stability. Issues arising from these evaluations will likely result in alternate control surface placement, size, or

configuration or result in changes to the vehicle center of gravity.

D. Sizing and Closure

Table 4 shows the fidelity definitions for the Sizing and Closure discipline. The closure methodology in IDEA

utilizes an “as drawn” and a scaled or “as closed” version of the vehicle geometry. Initially, the vehicle geometry is

defined at the “as drawn” level. For instance, as described previously, the second stage OML is defined by roughly

30 or so parameters, mostly physical dimensions of each of the main parts of the vehicle. Each of these parameters

contains a property that can be set by the user which determines whether that parameter is allowed to vary or not as

the vehicle is scaled. It also allows the user to specify the minimum and maximum allowable value for that

parameter. As closure begins, vehicle scaling is photographic (i.e. the scale factor in each primary axis is the same),

unless a scaling constraint is reached. For example, if a payload of fixed length, width and height is packaged, at

some point when photo-scaling down, continued scaling of the OML would result in the payload no longer fitting

within the vehicle, likely either in height or width. Here, scaling in that direction (height, width or both) would

cease, and scaling would continue in the unconstrained directions.

When the closure process begins, the “as drawn” and “as closed” vehicle scales are identical. In the simplest

form, closure is achieved by first computing the propellant fraction available (PFA) for the “as closed” vehicle.

Vehicle data (aerodynamic & propulsion databases, mass properties, etc.) are sent to the POST2 trajectory module

which flies the vehicle, optimizes on the given mission, and returns a propellant fraction required (PFR). The “as

closed” version of the geometry is then scaled up or down appropriately until PFA of the “as closed” equals PFR

from the POST2 run. When scaling reaches the point where PFA equals PFR, new vehicle data is generated for the

“as closed” version of the vehicle and passed to trajectory again for analysis. This cycle continues until convergence

is achieved, i.e. the PFA going into the trajectory analysis is the same (within some numerical tolerance) as the PFR

coming out. A similar closure process is currently being implemented for the first stage. For complete Level 1 and

Level 2 closure, the closure iterations would also include TPS and structural sizing based on the “as closed”

9


trajectory. Information from those sizing efforts will be included in the vehicle mass and volume updates on every

closure iteration.

E. Propulsion Design and Performance

For the Propulsion Design and Performance discipline, Table 5 shows the current analytical fidelity definitions.

Three main elements make up the tool suite for the area. For liquid rockets engines, IDEA utilizes a rocket

performance and sizing module built in AML by the U.S. Air Force and Technosoft that the Air Force uses in its

Reusable Military Launch System (RMLS) and Integrated Propulsion Analysis Tool (IPAT) environments3. This

module provides the user with the ability to select an existing engine from a database of over 40 predefined engines

or to create a new engine. This is accomplished by specifying some general sizing and performance information

about the engine or engines and selecting propellants from a list of nearly 40 fuels and seven oxidizers. The module

comes with mass estimating relationships based on physical dimensions and operating characteristics of the engine.

For scramjet engines, IDEA utilizes the SRGULL code22

, a tip-to-tail hypersonic cycle analysis tool developed

and used extensively at NASA Langley. SRGULL uses a two-dimensional Euler method for the forebody, inlet and

nozzle and a one-dimensional incremental combustor with an integral boundary layer method for all components. A

snapshot of the interface for the keel line design that has been developed in IDEA is shown in Figure 8. As

mentioned in the configuration section, the development of the

booster for the current TSTO vehicle begins with the definition of

the keel line. This interface allows the user to assemble the entire

high-speed keel line from scratch, using a building block approach.

Each of the flowpath components (forebody, inlet, isolator,

combustor, nozzle) are assembled from simple pieces of geometry

(e.g. lines, conics, circular arcs, etc.). Design rules are incorporated

that specify the relation between flowpath components. Once the

user is satisfied with the design, IDEA will generate the necessary

input data for SRGULL analyses and run the desired cases. For

optimization purposes, a design of experiments capability has also

been incorporated into the keel line design module.

For turbine analysis, plans are in place to integrate IDEA with

the Numerical Propulsion System Simulation (NPSS) tool from

NASA Glenn23

. NPSS has become an industry standard cycle

Figure 8. Snapshot of SRGULL / keel line

design interface in IDEA.

Table 4. Fidelity level definitions for the Sizing and Closure discipline.


Weight, volume and

dimensional closure

w/ consistent

bookkeeping of all

propellants, fluids

and other subsystems

needs, based on

commensurate

fidelity level inputs

from other

disciplines; All

outside analyses

input parameters

should be within +/-

30% of their final

closure values. “As

Closed” vehicle

photographic scale

factor < +/- 15%

from “As Drawn”

Weight, volume and

dimensional closure

w/ consistent

bookkeeping of all

propellants, fluids


needs, based on

commensurate


from other

disciplines; All

outside analyses

input parameters


20% of their final


Closed” vehicle

photographic scale

factor < +/- 10%

from “As Drawn”

Weight, volume and

dimensional closure

w/ consistent

bookkeeping of all

propellants, fluids


needs, based on

commensurate


from other

disciplines; All

outside analyses

input parameters


10% of their final


Closed” vehicle

photographic scale

factor < +/- 5% from

“As Drawn”

Weight, volume and

dimensional closure

w/ consistent

bookkeeping of all

propellants, fluids


needs, based on

commensurate


from other

disciplines; All

outside analyses

input parameters


5% of their final


Closed” vehicle

photographic scale


“As Drawn”

Weight, volume and

dimensional closure

w/ consistent

bookkeeping of all

propellants, fluids


needs, based on

commensurate


from other

disciplines; All

outside analyses

input parameters


2% of their final


Closed” vehicle

photographic scale


“As Drawn”

10


analysis tool in the turbomachinery world and is currently in use by all of the major aircraft engine manufacturers.

Efforts are also underway to develop a generic, high Mach number turbojet model in NPSS that will be driven by

IDEA.

Table 5. Fidelity level definitions for the Propulsion Design & Performance discipline.


Scaled empirical

estimates of engine Isp

and thrust-to-weight

ratio fixed based on

comparable engine

size, propellant

choices, and cycle

type. Engine vacuum

thrust scaled up or

down rubberised) to

meet requirements by

up to 50% while

holding selected Isp

and T/W

approximately

constant

1D cycle analysis

adjusted to level 2 or

higher results (MIL

standard or other

installation effects

included) Simple

steady-state

compustion chemistry

analysis based on

actual propellants,

mixture ratio, and

chamber pressure.

Determine vacuum Isp

based on actual

expansion ratio with

empirical corrections

from ideal expansion

solutions. Engine

thrust-to-weight ratio

to reflect actual nozzle

expansion ratio,

chamber pressure, and

chosen powerhead

cycle. Match "as-

analyzed" engine

vacuum thrust to

required thrust value

within 5% prior to

estimating vacuum Isp

and T/W.

2D/3D finite

difference inviscid

(Euler) flowfield

analysis w/ heat

conduction / transfer

& integral boundary

layer analysis.

Propulsive moments,

installation effects &

thermal balance

computed. Full power

balance for steady

state operation that

accurately represents

the selected power

cycle (matching pump

and turbine power).

Steady state

combustion chemistry

and nozzle flow to

predict Isp. Proper

accounting for gas

generator or

precombustor

performance, as

applicable. Empirical

estimates of nozzle

and chamber heat

transfer. Weight

modeling at the

component level (e.g.

individual pumps,

valves, lines, main

chamber, nozzle, etc.)

using individual

physics-based

estimating equations.

Engine not scaled

more than 1% from

"as-analyzed" engine

vacuum thrust before

analysis of Isp and

T/W must be re-

assessed.

2D/3D parabolized

Navier-Stokes finite

difference / volume

flowfield analysis w/

heat conduction /

transfer & integral

boundary layer

analysis. Propulsive

moments, installation

effects & thermal

balance computed.

Full mechanical

design. Full power

balance for both

steady state and

transient operation

(including

performance during

start-up and throttling

events). Combustion

chemistry at the level

of 3-D Navier-Stokes

CFD to account for

combustor and

injector performance,

nozzle flow (including

flow separation at low

altitude), and heat

transfer. Complete

modeling of

powerhead

components. Detailed

structural modeling to

predict engine weight

including FEA to

estimate weights of

major structural

components (chamber,

nozzle, lines,

turbomachinery, etc.).

Engine not scaled

more than 1% from

"as-analyzed" engine

vacuum thrust before

analysis of Isp and

T/W must be re-

assessed.

3D full or thin-layer

Navier-Stokes (FNS

or TLNS) flowfield

analysis including

pressure feedback,

shear stress & heat

transfer effects

computed directly.

Propulsive moments,

installation effects &

thermal balance

computed. Full

mechanical design.

Full power balance for

both steady state and

transient operation

(including start-up and

throttling events).

Combustion chemistry

at the level of 3-D

Navier-Stokes CFD to

directly account for

combustor and

injector performance,

nozzle flow (including

flow separation at low

altitude), and heat

transfer. Incorporate

component-level

hardware test data on

major elements such

as injectors and

pumps. Detailed

structural modeling to

predict engine weight

including FEA to

estimate weights of

major structural

components (chamber,

nozzle, lines,

turbomachinery, etc.)

under static as well as

dynamic loads.

Incorporate hardware

test data for predicting

engine component

weights. Engine not

scaled more than 0.5%

from "as-analyzed"

engine vacuum thrust

before analysis of Isp

and T/W must be re-

assessed.

11


F. Aerodynamics and Aerothermodynamics

Table 6 shows the fidelity definitions for the

Aerodynamics and Aerothermodynamics discipline. Tools

employed within this discipline vary based on the flight

condition being analyzed and the level of fidelity of interest.

At Level 1 for high-speed calculations (Mach > 3), the

IDEA environment will rely on SHABP24

and CBAero25

,

both of which have been integrated into IDEA, in order to

generate aerodynamic and heat transfer information. These

codes allow the user to choose from a variety of Newtonian

impact methods in order to estimate the lift, drag and

moment of the vehicle. A snapshot of an SHABP run at

hypersonic speeds on a representative first stage is shown in

Figure 9. For aerodynamic heating, both codes calculate the

location of streamlines along the vehicle surface and

estimate running lengths. Then, both estimate a skin friction

coefficient based on laminar or turbulent flow. Finally, they

calculate a convective heat transfer rate based on wall

temperature. SHABP allows a user-defined temperature profile (which is useful in analyzing hot structures) or will

calculate the profile based on a radiation equilibrium assumption. CBAero implements the radiation equilibrium

assumption. For low-speed aerodynamics, IDEA will employ the UDP slender body theory contained within

APAS26

. Similar theory is also available in CBAero and may be used as well.

Solutions from all of these codes will continuously be checked and updated with higher fidelity information

from a variety of CFD codes. For Level 2 aerodynamics, several options are currently under evaluation to support

the environment. The most likely candidate at this point is to use CART3D27

, an Euler code from NASA Ames,

although this option would require the addition of an integral boundary layer method to estimate viscous effects, in

accordance with the Level 2 fidelity definition. A design of experiments may also be employed in conjunction with

CART3D to reduce the required number of cases.

G. Thermal Management and TPS Sizing

The Level 0-4 analytic fidelity definitions for the Thermal Management and Thermal Protection System (TPS)

Sizing discipline are shown in Table 7. In the this discipline, plans are in place to incorporate the EXITS routine

from Miniver28

. EXITS is a finite element-based heat transfer code used for approximating transient temperature

distributions in one-dimensional (plug) models of TPS. Basic element groups, which model heat transfer based on

conductivity and capacitance of solids, radiation, convection within gases, and lumped mass thermal capacitance,

are utilized as building blocks for model construction and the assembly of any TPS concept. Pressure and

Figure 9. Snapshot of pressure coefficient

distribution for a hypersonic flight condition

generated by SHABP on the first stage.

Table 6. Fidelity level definitions for the Aerodynamics and Aerothermodynamics discipline.


Scaled

empirical

Linear/impact methods

with all drag

increments

(empirical)/Heating

(engineering-based)

adjusted to level 2 or

higher; vehicle

satisfies all

takeoff/landing speeds,

glide path, and runway

length requirements

(Including abort), no

control surface

deflections

3D CFD inviscid (Euler)

w/ integral boundary

layer or potential w/

semi-emperical drag

increments or thin layer

Navier Stokes w/ semi-

emperical non-viscous

drag increments, or CFD

anchored Level 1;

vehicle satifies all


glide path, runway

length, and longitudinal

stability requirements

(including abort)

3D CFD parabolized

Navier-Stokes (PNS)

finite difference /

volume flowfield

analsis w/ heat

conduction / transfer &

integral boundary layer

analysis; vehicle satifies

all takeoff/landing

speeds, glide path,

runway length, and

longitudinal, lateral &

yaw stability

requirements (Including

abort)

3D CFD full or thin layer

Navier-Stokes (FNS or

TLNS) flowfield analsis

including pressure

feedback, shear stress &

heat transfer efects

computed directly;

vehicle satifies all


glide path, runway

length, and longitudinal,

lateral & yaw stability

requirements (including

abort)

12


aerothermal heating, radiation to space, and convection to an

ambient temperature are used to define boundary conditions.

A capability has been added to IDEA to allow the user to

break the vehicle OML into regions for TPS sizing. A

example of the region definition capability is shown in

Figure 10. Outputs of the TPS sizing module include

material thickness distributions across the vehicle, as well as

mass estimates.

The TPS sizing capability will take advantage of the

structural load case generation capability mentioned

previously. Here, a set of sizing cases will be identified

along the trajectory profile that, when linearly interpolated,

will provide a basic reproduction of the flight profile

without having to analyze every trajectory time step. In

addition, future plans for the structural sizing module currently under development in IDEA include an extension

towards analysis of hot structures. A capability to analyze and appropriately size leading edges will also be included.

H. Airframe and Engine Subsystems

Table 8 shows the fidelity level definitions for the Airframe and Engine Subsystems discipline. In this area, the

plan for Level 1 is to implement existing MERs for all major engine and airframe subsystems. Currently,

EXAMINE has several sets of MERs that the user can choose, which include varying technology assumptions. At

present, many of these subsystem MERs have been incorporated into IDEA, although a comprehensive set is not yet

complete. In the long term, as dictated by the fidelity matrix in Table 7 for Level 2 analysis, more rigorous, physics-

based models of subsystems will be built that consider loading and environment information from the trajectory

simulation, plus thermal and power balance analyses. All of these influence the mass and volume of the individual

subsystem. Ideally, even more metadata will be tied or estimated for each system based on its characteristics, such as

technology level, failure rate, failure mode, maintenance requirements, etc. that can be used to feed life cycle

analyses estimates for the entire vehicle.

I. Life Cycle Analysis

For the Life Cycle Analysis disciplines, Table 9 shows the current fidelity definitions for reliability, operations

and cost. Several tools are under development or planned through NASA’s NRA process. Spaceworks Engineering

has just completed the development of a discrete event simulation (DES) of vehicle operations that estimates vehicle

characteristics such as turnaround time, maintenance requirements and operations cost. For development costs,

IDEA will utilize the NASA-Air Force Cost Model (NAFCOM) with updates to some of the cost estimating

Table 7. Fidelity level definitions for the Thermal Management and TPS Sizing discipline.


Parametric or

Historical

1D thru the thickness TPS

sizing for acreage; leading

edges evaluated with

assumed bluntness effects

to determine active /

passive requirement

Quasi-2D TPS

sizing for acreage;

blunt leading edges

analyzed; active

cooling rates

quantified

Quasi-2D TPS sizing for

acreage; blunt leading

edges analyzed; leading

edge cooling channels

sized; complete vehicle

thermal balance for flight

3D TPS sizing for

acreage; complete

vehicle thermal

balance including

ground and flight ops

Table 8. Fidelity level definitions for the Airframe and Engine Subsystems discipline.


Parametric or

Historical

Functional

definition &

evaluation and/or

1D or generic

modeling of

subsystem

Quantitative thermal &

fluid analysis of

subsystem; Component

masses estimated with

empirical and/or

historical data

Quantitative thermal,

fluid & power analysis of

subsystem; Component

masses estimated with

analytical data/analysis

Subsytem masses and

functional properties

based on actual

hardware specifications.

Figure 10. Example of TPS region definition

capability in IDEA that allows individual elements

to be grouped and sized together.

13


relationships (CERs) to better account for some of the airbreathing-specific elements of the vehicles of interest. Both

of these codes will be incorporated in Generation 2 of IDEA. As mentioned previously, the MDAO Discipline

Table 9. Fidelity level definitions for the Life Cycle Analysis disciplines.


Sa

fety

an

d R

eli

ab

ilit

y

Comparison to

historical systems or

quantification by limited expert

opinion

Reliability and Fault

Tree Analysis to the subsystem level with

historical failure rate

data; Preliminary documentation of

critical failure modes;

Deterministic or Probabilistic analysis

Reliability and Fault Tree

Analysis to the subsystem or

component level with historical or manufacturer failure rate data as

available; Limited accounting for

multiple failure modes (e.g. startup, dormant, continuous

operation) and common cause

failures; Probabilistic analysis according to MIL-STD-1629A or

equivalent; Probabilistic analysis

Formal (e.g. MIL-STD-217F or

equivalent) component level

bottoms-up reliability & safety assessment, FMECA, etc.;

Failure rate data from multiple

sources including existing databases, and manufacturers;

Full accounting for multiple

failure modes and common cause failures; Probabilistic

analysis; Quantitative FMECA

according to MIL STD-1629A or equivalent; Probabilistic

analysis

Component level bottoms-

up reliability &

safety assessment with

experimental

reliability test data;

Consideration

of formal program safety

plans, Hazard

Analysis

Fli

gh

t O

ps Comparison to

historical systems or

quantification by limited expert

opinion

Propulsion, TPS and other subsystems

estimated from aircraft

& space vehicle historical data &

adjusted for advanced

technology increments

Component-level maintainability

estimated from aircraft & space

vehicle historical data & adjusted for advanced technology

increments

Formal (e.g. MIL-STD-472 or

similar) maintainability assessment at the component

level; Maintenance data from

multiple sources including databases and manufacturers

Component level bottoms-

up O&M

assessment with experimental

maintenance

test data

Gro

un

d O

pera

tio

ns

Comparison to

historical systems or quantification by

limited expert

opinion

All ground processing

flow defined to the

facility level; Propulsion, TPS and

other major subsystems

estimated from aircraft & space vehicle

historical databases &

adjusted for advanced technology increments;

Deterministic analysis

with consideration of multiple scenarios

Discrete Event Simulation (DES) with events at the subsystem-level

(propulsion, TPS, structures);

Analysis of a single vehicle design with some design trades;

Probabilistic simulation; All

ground processing flow defined to the subsystem level. Subsystem

estimates based on historical

databases and adjusted for technology increments. Support

and processing time requirements

are based on analysis of a single point design with some trades

using probabilistic simulation.

DES at the component-level;

Broad design trade set exploration; Specific accounting

for resources (e.g. individual

facilities, technicians, etc.); Ground processing flow defined

to the component level. Specific

accounting for lowest level categories of resources.

Probabilistic simulation and

optimization

Support for Processing time

requirements

are based on component

failure rates and

requirements for both

nominal and

off-nominal processing.

Complete

resource and task-level

scheduling

Co

st

Weight based CER's

derived from aircraft or space vehicle

historical data with

adjustments for technology

complexity;

Deterministic analysis at the system

or subsystem level

Weight and programmatic factor

based CER's derived

from aircraft or space vehicle historical data

with adjustments for

technology complexity; Probabilistic analysis at

the subsystem level and

some component-level analysis

Component level parametric CER's; probabilistic analysis with

statistical accounting for

additional internal and external program risks/threats

Component or parts level

bottoms-up cost estimate with primary research of cost for

individual parts

Formal RFI

and/or actual

contractor bids

Eco

nom

ics

Basic financial

{break-even and net present value (NPV)}

assessment based on

supporting cost estimate and assumed

mission model;

market (demand) from from subset of

existing data;

Generally a single point estimate

Basic financial (break-

even and NPV)

assessment based on supporting cost estimate

and various mission

models; market (demand) from by

research of existing

market studies and projections; Multiple

scenarios considered

Financial estimate to include

aspects of capital structure such as

debt, equity, depreciation, etc.; Research of existing market

studies and projections; Market

possibly characterized by small customer survey or expert group;

Development of pro-forma

financial statements (income, balance sheet, cash flow);

Probabilistic analysis

Use of advanced financial analysis methods such as agent-

based simulation, options

analysis, and game theory; Development of formal business

plan and/or marketing plan for

commercial entities

Primary

research, consultation,

and quotes from

financial institutions,

regulatory

bodies, and investors /

stakeholders;

Market characterized

by focus

groups, large survey samples

14


would like to issue a topic area in an upcoming NRA call for improved safety and reliability models for hypersonic

vehicles. The call would contain elements for both models at the subsystem level, which could be integrated into

subsystem models under development in IDEA, as well as vehicle level methods for reliability estimation.

V. Schedule and Roadmap

The IDEA environment is being rolled out over several major milestones. Each generation of IDEA will build

upon the previous release. An overall schedule for the rollout is shown in Figure 11. Generation 0, which was

completed in FY09, provided the basic building blocks for performing vehicle closure. The POST2 trajectory

interface was combined with the sizing and closure algorithms and with the automated parametric packaging

capability so that a user could automatically resize and close the second stage for a given mission. Significant effort

was spent making the closure process robust, ensuring that from a wide range of initial inputs (vehicle dimensions,

propellant choices, mission parameters, etc.) the closure system was stable and would converge to a solution. The

final test for Generation 0 was an automated run of 117 design of experiments (DOE) cases that varied propellant

choice, staging conditions, payload mass, vehicle fineness and engine design parameters for the second stage. The

entire matrix was run, each starting from the same “as drawn” vehicle definition, without any failures, resulting in

closed vehicles with gross weights varying from 90,000 to 800,000 lbs and lengths from 55 to 127 feet. This

methodology is now being implemented on the first stage to achieve complete system closure.

Generation 1 will provide a complete Level 1 closure capability, incorporating all performance related

disciplines. Several of the modules under development are dependent on inputs and models from other disciplines

within the Hypersonics Project. The guidance, navigation and control discipline will be supplying methods for

stability and control evaluation, along with advanced, physics-based actuator sizing routines. The materials and

structures discipline is assisting with the TPS sizing routines, and the propulsion discipline is integrally involved in

the automated CFD meshing and lowspeed propulsion integration. As seen, the current timeline shows Generation 1

delivery in FY12. The Hypersonics Project has an Annual Performance Goal (APG), a Congressionally reported

milestone, related to the delivery of Generation 1 in FY12. According to the APG, to be completely successful,

Generation 1 will have to provide a complete re-closure of the TSTO concept in less than two days. As

demonstrated in the Joint System Study, the current timeframe for this capability is on the order of several months.

Figure 11. Schedule and major milestones for IDEA development and rollout.

15


While Generation 1 will complete the integration of the performance-related disciplines into IDEA, Generation

2 will increase the level of fidelity of these disciplines to Level 2. Generation 2 will also begin to integrate the life

cycle tools into the environment. As shown and previously discussed, several of these models are currently under

development through the NRA process, and several more are tentatively planned. Generation 2 will also begin to

expand on the vehicle classes that are included in IDEA. Higher fidelity analysis capabilities, such as analysis of hot

structures or structural dynamics models, will be included in Generation 3, as well as advanced optimization and

uncertainty methods.

VI. Summary and Conclusions

Hypersonic airbreathing systems, with their high level of integration and non-linear cross-discipline coupling,

demand that a multi-disciplinary approach be taken for their design, analysis and optimization. To solve this

problem, NASA’s Hypersonics Project is currently developing the Integrated Design and Engineering Analysis

(IDEA) environment. IDEA is a collaborative environment for parametrically modeling conceptual and preliminary

launch vehicle configurations using the Adaptive Modeling Language (AML) as the underlying framework. The

environment integrates geometry, packaging, subsystems, propulsion, aerodynamics, aerothermodynamics,

trajectory, closure, and structural analysis into a generative, parametric, unified computational model where data is

shared seamlessly between the different disciplines. A matrix of various fidelity levels for each of these disciplines

has been introduced. IDEA environment development is currently being focused on mid-level fidelity analyses, i.e.

those that should be sufficient to bring a concept to a System Requirements Review phase of a project. Substantial

progress has been made in the development. The first version of the environment, Generation 0, has already been

completed, and work is well underway on Generation 1, due to be delivered in FY12.

References

1 Moses,P.L., Bouchard,K.A., Vause,R.F., Pinckney,S.Z., Ferlemann,S.M., Leonard,C.P., Taylor,L.W.III,

Robinson,J.S., Martin,J.G., Petley,D.H. and Hunt,J.L.; Design and Development of an Airbreathing Launch Vehicle

with Turbine-Based Lowspeed and Dual Mode Ramjet / Scramjet Highspeed Propulsion; presented at the 36th

JANNAF CS/APS/PSHS Joint Meeting, Cocoa Beach, FL; October 1999. 2 Responding to the Call: Aviation Plan for American Leadership; published by the National Institute of

Aerospace, Hampton, VA; 2005. (http://www.nianet.org/pubs/AviationPlan.php) 3 Zweber,J.V., Kabis,H., Follett,W.W., and Ramabadran,N.; Towards an Integrated Modeling Environment for

Hypersonic Vehicle Design and Synthesis; AIAA-2002-5172; September 2002. 4 Bowcutt,K.G., Kuruvila,G., Grandine,T.A., Hogan,T.A., and Cramer,E.J.; Advancements in Multidisciplinary

Design Optimization Applied to Hypersonic Vehicles to Achieve Closure; AIAA-2008-2591; April 2008. 5 Robinson,J.S., Martin,J.G., Bowles,J.V., Mehta,U.B., and Snyder,C.A.; An Overview of the Role of Systems

Analysis in NASA’s Hypersonics Project; AIAA-2006-8013; November 2006. 6 Monell,D., Mathias,D., Reuther,J., and Garn,M.; Multi-Disciplinary Analysis for Future Launch Systems

Using NASA’s Advanced Engineering Environment (AEE); AIAA-2003-3428; June 2003. 7 The Adaptive Modeling Language, TechnoSoft, Inc., Cincinnati, OH; (http://www.technosoft.com/)

8 Ferlemann,S.M., Robinson,J.S., Leonard,C.P., Taylor,L.W.III, Martin,J.G. and Kamhawi,H.; Developing

Conceptual Hypersonic Airbreathing Engines Using Design of Experiments Methods; AIAA-2000-2694; June 2000. 9 Rehder, J.J.; Advanced Vehicle Integration & Synthesis Environment (AdVISE); presented at the First Annual

Space Systems Engineering Conference, November 8-10, 2005, Atlanta, Georgia 10

Bilardo,V.J., Robinson,J.S., Komar,D.R., Taylor,W., Lovell,N.T. and Maggio,G.; NGLT Systems Analysis

Tiger Team Results: Implications for Space Launch Architectures and Technologies; AIAA-2003-5263; July 2003. 11

NASA Procedural Requirement (NPR) 7120.5D; NASA Space Flight Program and Project Management

Requirements; NASA Office of the Chief Engineer. 12

Hammond,W.E.; Design Methodologies for Space Transportation Systems; AIAA Educational Series, 2001. 13

Komar,D.R., Hoffman,J., Olds,A., and Seal,M.; Framework for the Parametric System Modeling of Space

Exploration Architectures; AIAA-2008-7845; September 2008. 14

DePasquale,D., Kelly,M., Charania,A.C., and Olds,J.; Activity-Based Simulation of Future Launch Vehicle

Ground Operations; AIAA-2008-7660; September 2008. 15

Ferlemann,P. and Kamhawi,H.; Automated Structured CFD Mesh Generation in NASA’s IDEA Environment;

presented at the JANNAF Modeling & Simulation Meeting, December 2008.

16


16

White, J.A., and Morrison, J.H.; A Psuedo-Temporal Multi-Grid Relaxation Scheme for Solving the

Parabolized Navier-Stokes Equations; AIAA-99-3360; July 1999. 17

Ferlemann,P.; Hyper-X Mach 10 Scramjet Preflight Predictions and Flight Data; AIAA-2005-3352; May

2005. 18

Meyer,B.; A Comparison of Computational Results with Transonic Test Data from a Hypersonic

Airbreathing Model with Exhaust Simulation; AIAA-2003-4411; July 2003. 19

Bruhn, E.F.; Analysis and Design of Flight Vehicle Structures; Revised Edition, June 1973. 20

Eldred,L. and Kamhawi,H.; Automated Structural Analysis in NASA’s IDEA Environment; presented at the

JANNAF Modeling & Simulation Meeting, December 2008. 21

Striepe, S.A., Powell, R.W., Desai, P.N., Queen, E.M., Brauer, G.L., Cornick, D.E., Olson, D.W., Petersen,

F.M., Stevenson, R., Engel, M.C., Marsh, S.M., and Gromoko, A.M.; Program to Optimize Simulated Trajectories

(POST II), Vol. II Utilization Manual Version 1.1.6.G; January 2004, NASA Langley Research Center, Hampton,

VA. 22

Pinckney, S.Z. and Walton, J.T.; Program SRGULL: An Advanced Engineering Model for the Prediction of

Airframe-Integrated Subsonic/Supersonic Hydrogen Combustion Ramjet Cycle Performance; NASP TM-1120,

January 1991. 23

Evans, A.L., etal; Numerical Propulsion System Simulation's National Cycle Program; AIAA-98-3113, 1998. 24

Gentry, A. E., Smyth, D. N., Oliver, W. R.; The Mark IV Supersonic-Hypersonic Arbitrary-Body Program;

McDonnell-Douglas Corporation, Technical Report AFFDL-TR-73-159, November 1973. 25

Kinney, D.J.; Aero-Thermodynamics for Conceptual Design; AIAA-2004-31, January 2004. 26

Sova, G., Divan, P.; Aerodynamic Preliminary System II; North American Aircraft Operations, Rockwell

International, Los Angeles, CA. 27

Aftosmis, M.J., Berger, M.J., and Adomavicius, G. D.; A Parallel Multilevel Method for Adaptively

Refined Cartesian Grids with Embedded Boundaries; in Proc. 38th AIAA Aerospace Sciences Meeting &

Exhibit, AIAA-2000-0808, Reno, NV, Jan. 2000. 28

Engel, C. D., and Praharaj, S. C.; MINIVER Upgrade for the AVID System, Vol. I: LANMIN User’s Manual;

NASA CR-172212, 1983.

x-43d conceptual design and feasibility study - nasa technical

Documents